IBP01101

PIPELINE MECHANICAL DAMAGE: DETECTION,

ASSESSMENT AND MONITORING USING THE SAAM
PIPELINE INSPECTION TOOL
David A. Russell1, Gordon C. Short2
Copyright 2001, Brazilian Petroleum and Gas Institute - IBP
This paper was prepared for presentation at the 3rd Seminar on Pipeline, held in Rio de Janeiro, Brazil, 21-23 November, 2001
This paper was selected for presentation by the Event Technical Committee following review of information contained in an abstract submitted by the
author(s). Contents of the paper, as presented, have not been reviewed by the IBP. Organizers will neither translate nor correct texts received. The
material, as presented, does not necessarily reflect any position of the Brazilian Petroleum Institute, its officers, or members.

Abstract

This paper presents the results of work carried out by RST Projects Limited using the SAAM
pipeline inspection technology to detect, assess and monitor mechanical damage and changes in the
internal bore of hydrocarbon pipelines.

One of the most important uses of SAAM has been to detect, locate and monitor mechanical
damages and changes in the internal bore of hydrocarbon pipelines. This paper explains how
SAAM can be used in this role and specifically the types of features which can be detected. These
include dents, ranging from those which are so severe as to cause damage to the carrier pig through
relatively small features, as well as flat spots caused by field bending and manufacturing defects.
Examples of each are presented. A discussion of how this technology can be used as a routine
condition-monitoring tool to provide early warning of the presence of mechanical damage is also
included.

The paper draws upon experiences gained on projects in the UK, Middle East and Africa and
wherever possible makes use of case studies.

Introduction

RST Projects Limited, a Scottish pipeline inspection company, has developed the SAAM
technology, an innovative system that allows the internal inspection of pipelines while they are
cleaned. The development is a result of many years of experience in the pipeline industry and
research into how internal variables within a pipeline can affect the behaviour of cleaning pigs. The
aim was to develop a black box flight recorder which when fitted on board a traditional cleaning
pig would measure changes in the dynamics of the pig as it travels through the line. The result, after
an initial 3 years R&D programme, was the SAAM (Smart Acquisition Analysis Module) system,
a cost effective, low risk and easy to use pipeline inspection technology. SAAM was deployed
operationally for the first time in July 1997.

1
Managing Director, RST Projects, Edinburgh, UK
2
Technical Director, RST Projects, Edinburgh, UK
1
 IBP01101

1 - Background

The safe and continuous operation of hydrocarbon pipelines requires the pipeline operator to have
the most up-to-date information to hand regarding the condition of his pipeline. It is clear from a
number of surveys that the most common cause of pipeline failure is third-party external damage,
caused by e.g. ploughing of buried lines, or the impact of trawl gear against (unburied) sub-sea
lines. Furthermore, in many areas in the world, the effects of sabotage or vandalism must also be
considered.

For example, according to the US Department of Transport, [1], of the more than 900 pipeline
failure incidents in the continental United States in the years 1997-2000, around a third were due to
third-party damage incidents. A major study of pipeline leakages, undertaken in 1998 by Concawe
confirmed this figure.

In locations where the pipeline is relatively inaccessible, the obvious standby of visual inspection is
not an easy option to implement, and quite significant damage may go undetected for considerable
periods of time. The continued operation of upstream oil and gas pipelines in these often rather
inaccessible locations requires the ability to monitor the line condition remotely. In many cases,
there will be no simple way to detect even relatively gross line damage by means of an exterior
survey. The use of semi-intelligent or Smart Utility Pigs for these applications seems to offer the
best way forward.

These smart utility pigs have been developed during recent years to perform a variety of functions
with the minimum disruption to operations. RST Projects SAAM tools are a good example of the
capabilities of these types of tool. They are able to record temperature, pressure (both absolute and
differential), inclination (acceleration) and vibration. Other instrumentation can be integrated with
the data recorder as required for
specialist applications. These devices
have found a number of applications
in the North Sea and other upstream
oil and gas regions, such as the
detection and monitoring of pipeline
out-of-straightness features (mapping)
[2], the identification, location and
monitoring of dents, status of in-line
components, internal anomalies and
zones of wax formation. Other
applications include logging of
pipeline process data, diagnosis of
pigging problems, and facilitating the
optimisation of pigging programmes.

Figure 1 - SAAM Unit in Pig

2
 IBP01101

The essential philosophy behind the use of these tools is to highlight changes that may be occurring
in the pipeline by the use of regular surveys. This methodology is made practical by the relatively
low per kilometre survey cost, and the potential to incorporate these inspections into a routine
cleaning operation. Figure 1 shows a typical way in which a smart utility pig can be mounted in a
standard cleaning pig. Current versions can be deployed in lines down to 6" diameter, and survey
transit times of up to 14 days are achievable. The SAAM technology is now fully field proven with
some 3,000km of pipelines having been surveyed so far. The technology aims at facilitating the
condition monitoring and decision-making process for pipeline operators.

It is the purpose of this paper to discuss a number of situations in which these techniques have
proven useful for the detection of pipeline damage.

2 - Basis Of Technique For Detection Of Pipeline Out-Of-Roundness

Consideration needs to be given to the mechanisms by which pipeline damage can be detected by
the instrumentation onboard a standard SAAM unit. Considering the motion of a pig through a
pipeline, it is obvious that the main effect is the interaction between the pig disks and the pipe-wall.
The obvious outcome of this is that the driving differential pressure necessary to move the pig down
the pipeline depends upon this interaction. Thus this value is in some sense at least characteristic of
a given pig in a given line.

Vibration is also experienced by all pigs to a greater or lesser extent as they travel down a line. If
the disk/wall contact is well lubricated and motion is steady, as for a large diameter pig in an oil
line, the normal level of vibration will be quite low. For poorly lubricated motion (gas or water),
especially at low velocities, were the pig is more likely to experience stick-slip motion, background
levels of vibration can be very high. It has also been observed that spirally-wound line pipe also
tends to give rise to higher vibration levels.

The vibration is measured by RST's

Figure 2 - Vibration Signal at Weld
Smart Utility Pig in a direction co-axial
to the motion of the pig. In normal use
this means that the only source of major
disturbances in the vibration signal is the
'judder' as the pig passes over each weld.
This has a characteristic double peak
structure as a bi-di pig passes a weld, as
each set of disks (front and rear) traverse
the weld. Figure 2 shows a typical signal.
Under some circumstances (slow motion)
it is possible to distinguish further
structure due to the seal and guide disks
in each set. These weld kicks are the
principal means by which the progress of
the pig down the line is monitored.

3
 IBP01101

A third detectable effect is the tendency of the pig to hold up at obstructions, visible as an
acceleration feature, where the pig stops and then sets off again. This type of effect is occasionally
seen at welds in oil pipelines, particularly where the pig is 'tight', but is much more common in gas
lines, where the compressibility of the line fluid allows momentary fluctuations in pig sped much
more readily.

During RST's experience in running inspections with these tools, a wide range of different
anomalies have been observed in the full range of pipelines. The following types of anomaly will be
considered in detail below: changes in pipe specification; flats spots and out-of roundness; and
dents and mechanical damage.

3 - Detection Of Changes In Pipeline ID

All this leads naturally to the consideration of the effects on these readings caused by restrictions in
the line. The main effect caused by mechanical damage which is visible in SAAM data is the
increased resistance to the motion of the pig. This is manifested in a number of different ways
which are detectable by instrumentation on board a Smart Utility Pig. First and most obviously, a
significant obstruction, particularly any reduction in internal bore or out-of-roundness will cause a
rise in differential pressure as the pig passes the obstruction.

As an example, the case of a simple

change in line specification from
normal to heavy-wall pipe may
involve an overall change in ID (for
OD controlled line sizes) of only a
few millimetres. For pigs for which
the disks are sized to be suitable for
the larger internal diameter, this type
of spec change can cause the driving
DP to increase by 50% or more. An
example of this is Figure 3, showing
the top of a pipeline riser in the UK
sector of the North Sea. The figure
shows the driving differential pressure
increasing from 0.10-0.15 bar in the
normal part of the line to 0.25-0.30
Figure 3 - Differential Pressure in Pipeline Riser
bar in the heavy wall pipe at the
splash zone and above. Similarly
Figure 4 shows pressure/vibration signals for a transition from heavy wall into standard line pipe for
a water-driven pig in an 18" line. The change in ID here is 6mm, and again the DP is almost twice
as high in the heavy-wall section. In this case it was fairly clear that the pig's guide disks, which
were sized for the larger diameter part of the line, were interfering with pipe-wall in the heavy-
walled section. It is this that is responsible for the significant increase in vibration also apparent in
the figure.

4
 IBP01101

The most important conclusion is that a relatively small change in pipeline internal diameter can
cause a major change in the differential pressure and vibration characteristic recorded by a Smart
Utility Pig.

4 - Detection Of Flat Spots In

Pipelines

The previous section covered the

characteristics of known (and
planned) changes in the internal bore
of a pipeline. While this can be of
interest under certain circumstances,
the more general problem of detecting
shorter, unplanned changes in the
internal bore, both in terms of out-of-
roundness features such as ovality or
flat spots, and in terms of dents and
mechanical damage, has much greater
application.
Figure 4 - Differential Pressure at ID Change

Considering flats spots first, these can be expected to be very similar to changes in pipe
specification, the difference being that the magnitude of the effects on the differential pressure and
vibration characteristics are expected to be much smaller.

Figures 5 and 6 show two examples of flat spots from a 32" oil export line. This line lies onshore,
Figure 5 - Example of Pipeline Flatspot
and has numerous bends of a few
degrees where the pipe was cold-bent
on site during construction. In Figure
5, the feature shows up as an increase
in DP from 0.35 bar to 0.45 bar. This
is a rise of 50% rather than 100% or
more, as was the case for the overall
change in diameter. Figure 6 shows a
slightly larger feature, with and
increase in DP from 0.3 to 0.5 bar. In
both these cases the accelerometer
inclination trace shows that this peak
coincides with a downward bend by a
few degrees. In this particular pipeline
the operator was aware that the
process by which these bends were
formed was likely to have resulted in
the formation of a flat spot on the
outer side of the bend.

5
 IBP01101

In both cases the vibration also shows a peak at the same location. Perhaps the most important
characteristic to note here is the relatively smooth DP profile.

5 - Dents And Other Out-Of-

Roundness Features

Dents caused by actual impacts

against the pipeline differ from the
types of features discussed above in
that the beginning and end of the
feature will tend to be much more
sudden. Figure 7 shows an example of
this from the same pipeline discussed
in the previous section. Here the peak
in DP, although comparable in size to
the features caused by simple out-of-
roundness, rises much more sharply.
Interestingly the peak in vibration is
much shallower than that in the out-
of-roundness. This dent is known to
be on the underside of the pipe, and to Figure 6 - Example of Pipeline Flatspot
be relatively shallow.

For comparison, Figure 8 shows the

effect of a much more severe dent.
Here the pig is significantly
obstructed by a feature, known to be
on the top of the line, and to be
sufficiently deep to interfere with the
pig body in an 18" line. This figure
shows first the pig running into the
dent, with the driving pressure rising
to around 0.8 bar. The pig then runs
into the part of the line which was
sufficiently damaged to interfere with
the pig body. It is possible to see the
pig being forced nose down in the
inclination trace.

Figure 7 - Example of Dent

6
 IBP01101

Figure 9 shows the state of the carrier

pig after impact with the restriction.

6 - Limitations Of Technique

There are still a number of obvious

limitations to the use of this technique
at present. The most obvious of these
is that at present only an indication of
the presence of dents and other out-
of-roundness features can be obtained,
along with perhaps a qualitative
measure of their size. No information
on the absolute size of features is Figure 8 - Example of Pipeline Obstruction
generated, nor of their position on the
pipe's circumference.

Other issues that still require to be dealt with include the problems of detecting dents which happen
to be close to welds. The vibration signature in this case will be obscured, and although it is not
normal to see peaks in differential pressure at welds, this behaviour has been observed to occur. In
particular this is thought to be present where there is significant (excessive) weld penetration. This
problem can be tackled in many cases by using the regularity of the pattern of the weld vibration
trace to 'pre-whiten' the vibration time series, in effect removing the welds. Anomalies in the
vibration (and DP) characteristics should then show up much more clearly.

Disk wear too, may cause difficulties

in comparing like with like, even
within the same pipeline and the same
survey, if conditions are such that
considerable wear occurs during a
single pigging run.

The biggest obstacle to the successful

Figure 9 - Damaged Pig
detection of mechanical damage
features in a pipeline using smart
utility pigs such as SAAM lies in the
difficulty of amassing sufficient
information on known features, across
a wide range of lines sizes and flow
conditions.

7
 IBP01101

7 - Future Developments

The obvious next step is to advance the extent to which these types of anomaly can be recognised
and even measured using these Smart Utility Pig techniques. RST Projects are currently engaged in
a test programme which will completely characterise the pig's behaviour at a restriction in a test
situation. Together with a growing database of characteristics for pig behaviour this will provide a
route to the validation of the SAAM tool's ability to reliably detect dents and other anomalies. This
will allow confidence limits to be developed as to the smallest features (in both length and depth)
which can be detected, and begin to put a quantitative calibration of the tool in place.

It should be fairly clear that, particularly when the normal motion is as smooth as shown in Figure
2, that any slight hindrance to the pig's passage is likely to be evident as a disturbance in the
vibration trace. This is, in itself, not evidence of mechanical damage, or even of a defect. In any
given single pig run, it is perfectly conceivable that dynamic effects, associated with the motion of
the pig, could cause a similar disturbance, which, after all, causes an unsteady flow situation in the
line during the pigging run.

On the other hand, if repeated surveys of the same line come up with the comparable results, the
cause must obviously be sought in some local defect in the line. This has the weakness that only
inferential evidence is available regarding possible damage, however, by comparing data traces with
those caused by known types of defect it is possible to obtain at least qualitative information
regarding the line condition. In the case of significant out-of-roundness, where there is some
reasonably large part of the pipeline's circumference that is distorted, the effects on DP will be very
similar to those described above. The magnitude of the DP increase, of course may be significantly
less.

This is the major benefit of the smart utility pig approach. The low cost of incorporating an
inspection capability into a routine pipeline cleaning programme, by fitting inspection technology
inside a standard cleaning pig allows the frequent monitoring of pipeline conditions, and gives
early warning of when damage has occurred.

References

1. US DEPARTMENT OF TRANSPORTATION, Office of Pipeline Safety.

http://ops.dot.gov/stats.htm, Web site, 2001.
2. SHORT, Gordon, OGUNJIMI, Patrick. Out-of-straightness detection, assessment, and
monitoring using the SAAM pipeline-inspection tool. In: 13th International Pipeline Pigging,
Integrity Assessment & Repair, 2001, Houston.